December 8, 2016
Whether you’re completely new to the PPC industry or you’re an expert, there are always questions around what CRO is and why it’s important. I’ll walk you through what CRO is, what kind of testing you can do to optimize and why this is so important to your PPC account.
What is CRO?
CRO stands for conversion rate optimization. Yes, we want to optimize for conversion rate. However, this isn’t the only thing you should focus on. When you’re making optimizations, be sure you’re keeping in mind the main metrics your company needs to track. What are your goals? This could be conversion rate, revenue, average order value, phone calls, lifetime customer value, etc.
What Does Conversion Rate Optimization (CRO) involve?
Testing. Having a successful CRO program requires testing, testing and more testing. You can form opinions and hypotheses about your site but you need to look to the data for the answers. Why can’t we just implement commonalities or best practices we see across the industry to our website? First, best practices don’t always work. Check out Kate Wilcox’s post to find out why. Second, your users are going to perform and act differently than your competitor’s users. You need to test your changes before implementation to ensure you’re making the right decision for your users.
Determining What To Test
It’s all about your users. What do they need to better their experience? So many people just throw a test against the wall and hope it sticks. Don’t be surprised when results are inconclusive.
There needs to be reason behind every test if you want to develop a successful CRO program.
- Analytics can help you determine where to test. Are there problem pages? Is mobile tanking your conversion rate? Are your users returning and still not converting?
- PPC Account
- Many people might not think to take a step back and look where your paid search traffic is coming from. Are expectations being set in the ads that aren’t being met? Are we even sending traffic to the most appropriate page?
- Heat mapping
- Analyze heat maps to determine where your users are clicking and looking throughout the page. You’ll be able to see where the focus currently is and determine how to focus the attention where it needs to be.
- User Recordings
- Analyze your actual users as well as users who have never seen your site before. You’ll learn problem areas, see what they expect as they navigate the site and how those expectations are or are not met.
Using a combination of these tools and analyzing your findings will really help determine what your users need and how you can incorporate this into your testing schedule. Find the problem. Hypothesize a solution. Test this hypothesis.
How do I test?
You need to test your landing pages to ensure you’re giving users exactly what they need to make a decision. There are a few different types of tests you can run.
- A/B Tests
- An A/B test involves one change to your page. This could be a number of different things. For example, you could change the headline, add testimonials to a page, increase or decrease the number of fields in a form, add a form, remove navigation, etc. There are so many options but you do want to be sure you limit this to one change so you can correctly attribute the results to that change. If you change more than one thing in a test, you ultimately contaminate the results and have no way to determine which of these changes influenced your users.
- Multivariate Tests
- Multivariate tests involve modifications to multiple variables. If you want to determine which combination of variations performs the best, you’ll use multivariate testing. For example, if you want to test which call-to-action works best in a hero image as well as the language in the sub-header, you can create a multivariate test to see which combination of sub-header text and call-to-actions work the best.
- Split URL Tests
- A split URL test involves two URLs. There are a couple of ways to approach this type of test. The first requires the use of your PPC account. You may want to see if paid search traffic from a certain campaign would convert more seeing a different page. You can easily run the split URL test for specific campaigns to determine which page is best fit for that campaign. The second is if you’re testing a different layout to the page. You might have a short and concise page but want to see if providing your users with a longer, more detailed page will lead them to convert. Maybe you have a lengthy, distracting page and want to test a shorter page that is to the point and only gives users the necessary information.
An important thing to note is that there are testing platforms out there to make setting up these tests easy: Visual Website Optimizer (VWO), Optimizely and Unbounce just to name a few. VWO & Optimizely will allow you to run the tests we’ve described. Unbounce will allow you to create an entirely new landing page and continuously test this new page.
Why is CRO so Important?
You can drive all the traffic in the world to your site but if your site is broken, if your users get frustrated, if it’s hard to navigate, you end up with wasted spend and missed conversions. When you’re looking to increase conversions, don’t assume you need to increase your traffic from paid search to do so. Rather than increasing your spend, utilize a CRO program to convert the traffic you’re already sending to the site.
See our latest webinar for more information on A/B testing, how long to run and when to conclude these tests, Eisenberg’s Hierarchy of Optimization along with some of our examples of winning and losing tests.